/* -*- c++ -*- */
/*
 * Copyright 2015,2016,2018,2019 Free Software Foundation, Inc.
 *
 * SPDX-License-Identifier: GPL-3.0-or-later
 *
 */

#ifdef HAVE_CONFIG_H
#include "config.h"
#endif

#include "dvbt2_interleaver_bb_impl.h"
#include <gnuradio/io_signature.h>

namespace gr {
namespace dtv {

dvbt2_interleaver_bb::sptr dvbt2_interleaver_bb::make(dvb_framesize_t framesize,
                                                      dvb_code_rate_t rate,
                                                      dvb_constellation_t constellation)
{
    return gnuradio::make_block_sptr<dvbt2_interleaver_bb_impl>(
        framesize, rate, constellation);
}

/*
 * The private constructor
 */
dvbt2_interleaver_bb_impl::dvbt2_interleaver_bb_impl(dvb_framesize_t framesize,
                                                     dvb_code_rate_t rate,
                                                     dvb_constellation_t constellation)
    : gr::block("dvbt2_interleaver_bb",
                gr::io_signature::make(1, 1, sizeof(unsigned char)),
                gr::io_signature::make(1, 1, sizeof(unsigned char)))
{
    signal_constellation = constellation;
    code_rate = rate;
    if (framesize == FECFRAME_NORMAL) {
        frame_size = FRAME_SIZE_NORMAL;
        switch (rate) {
        case C1_2:
            nbch = 32400;
            q_val = 90;
            break;
        case C3_5:
            nbch = 38880;
            q_val = 72;
            break;
        case C2_3:
            nbch = 43200;
            q_val = 60;
            break;
        case C3_4:
            nbch = 48600;
            q_val = 45;
            break;
        case C4_5:
            nbch = 51840;
            q_val = 36;
            break;
        case C5_6:
            nbch = 54000;
            q_val = 30;
            break;
        default:
            nbch = 0;
            q_val = 0;
            break;
        }
    } else {
        frame_size = FRAME_SIZE_SHORT;
        switch (rate) {
        case C1_3:
            nbch = 5400;
            q_val = 30;
            break;
        case C2_5:
            nbch = 6480;
            q_val = 27;
            break;
        case C1_2:
            nbch = 7200;
            q_val = 25;
            break;
        case C3_5:
            nbch = 9720;
            q_val = 18;
            break;
        case C2_3:
            nbch = 10800;
            q_val = 15;
            break;
        case C3_4:
            nbch = 11880;
            q_val = 12;
            break;
        case C4_5:
            nbch = 12600;
            q_val = 10;
            break;
        case C5_6:
            nbch = 13320;
            q_val = 8;
            break;
        default:
            nbch = 0;
            q_val = 0;
            break;
        }
    }
    switch (constellation) {
    case MOD_QPSK:
        mod = 2;
        set_output_multiple(frame_size / mod);
        packed_items = frame_size / mod;
        break;
    case MOD_16QAM:
        mod = 4;
        set_output_multiple(frame_size / mod);
        packed_items = frame_size / mod;
        break;
    case MOD_64QAM:
        mod = 6;
        set_output_multiple(frame_size / mod);
        packed_items = frame_size / mod;
        break;
    case MOD_256QAM:
        mod = 8;
        set_output_multiple(frame_size / mod);
        packed_items = frame_size / mod;
        break;
    default:
        mod = 1;
        set_output_multiple(frame_size / mod);
        packed_items = frame_size / mod;
        break;
    }
    generate_lookup();
}

inline void dvbt2_interleaver_bb_impl::interleave_parity_bits(int* tempu, const int*& in)
{
    for (int k = 0; k < nbch; k++) {
        tempu[k] = *in++;
    }
    for (int t = 0; t < q_val; t++) {
        for (int s = 0; s < 360; s++) {
            tempu[nbch + (360 * t) + s] = in[(q_val * s) + t];
        }
    }
    in = in + (q_val * 360);
}

inline void dvbt2_interleaver_bb_impl::twist_interleave_columns(
    int* tempv, int* tempu, int rows, int mod, const int* twist)
{
    int index = 0, offset;
    for (int col = 0; col < mod; col++) {
        offset = twist[col];
        for (int row = 0; row < rows; row++) {
            tempv[offset + (rows * col)] = tempu[index++];
            offset++;
            if (offset == rows) {
                offset = 0;
            }
        }
    }
}

void dvbt2_interleaver_bb_impl::generate_lookup()
{
    int rows, index = 0;
    int* tempv;
    int* tempu;
    const int* twist;
    const int *c1, *c2, *c3, *c4, *c5, *c6, *c7, *c8;
    const int *c9, *c10, *c11, *c12, *c13, *c14, *c15, *c16;

    tempv = new int[FRAME_SIZE_NORMAL];
    tempu = new int[FRAME_SIZE_NORMAL];

    for (int i = 0; i < FRAME_SIZE_NORMAL; i++) {
        lookup_table[i] = i;
    }

    const int* in = &lookup_table[0];

    switch (signal_constellation) {
        // ignore QPSK, not worth it, as there is no column interleaving
    case MOD_16QAM:
        rows = frame_size / (mod * 2);

        if (frame_size == FRAME_SIZE_NORMAL) {
            twist = &twist16n[0];
        } else {
            twist = &twist16s[0];
        }

        interleave_parity_bits(tempu, in);

        c1 = &tempv[0];
        c2 = &tempv[rows];
        c3 = &tempv[rows * 2];
        c4 = &tempv[rows * 3];
        c5 = &tempv[rows * 4];
        c6 = &tempv[rows * 5];
        c7 = &tempv[rows * 6];
        c8 = &tempv[rows * 7];

        twist_interleave_columns(tempv, tempu, rows, mod * 2, twist);

        for (int j = 0; j < rows; j++) {
            tempu[index++] = c1[j];
            tempu[index++] = c2[j];
            tempu[index++] = c3[j];
            tempu[index++] = c4[j];
            tempu[index++] = c5[j];
            tempu[index++] = c6[j];
            tempu[index++] = c7[j];
            tempu[index++] = c8[j];
        }
        break;

    case MOD_64QAM:
        rows = frame_size / (mod * 2);

        if (frame_size == FRAME_SIZE_NORMAL) {
            twist = twist64n;
        } else {
            twist = twist64s;
        }

        interleave_parity_bits(tempu, in);

        c1 = &tempv[0];
        c2 = &tempv[rows];
        c3 = &tempv[rows * 2];
        c4 = &tempv[rows * 3];
        c5 = &tempv[rows * 4];
        c6 = &tempv[rows * 5];
        c7 = &tempv[rows * 6];
        c8 = &tempv[rows * 7];
        c9 = &tempv[rows * 8];
        c10 = &tempv[rows * 9];
        c11 = &tempv[rows * 10];
        c12 = &tempv[rows * 11];

        twist_interleave_columns(tempv, tempu, rows, mod * 2, twist);

        for (int j = 0; j < rows; j++) {
            tempu[index++] = c1[j];
            tempu[index++] = c2[j];
            tempu[index++] = c3[j];
            tempu[index++] = c4[j];
            tempu[index++] = c5[j];
            tempu[index++] = c6[j];
            tempu[index++] = c7[j];
            tempu[index++] = c8[j];
            tempu[index++] = c9[j];
            tempu[index++] = c10[j];
            tempu[index++] = c11[j];
            tempu[index++] = c12[j];
        }
        break;

    case MOD_256QAM:
        if (frame_size == FRAME_SIZE_NORMAL) {
            rows = frame_size / (mod * 2);

            interleave_parity_bits(tempu, in);

            c1 = &tempv[0];
            c2 = &tempv[rows];
            c3 = &tempv[rows * 2];
            c4 = &tempv[rows * 3];
            c5 = &tempv[rows * 4];
            c6 = &tempv[rows * 5];
            c7 = &tempv[rows * 6];
            c8 = &tempv[rows * 7];
            c9 = &tempv[rows * 8];
            c10 = &tempv[rows * 9];
            c11 = &tempv[rows * 10];
            c12 = &tempv[rows * 11];
            c13 = &tempv[rows * 12];
            c14 = &tempv[rows * 13];
            c15 = &tempv[rows * 14];
            c16 = &tempv[rows * 15];

            twist_interleave_columns(tempv, tempu, rows, mod * 2, twist256n);

            for (int j = 0; j < rows; j++) {
                tempu[index++] = c1[j];
                tempu[index++] = c2[j];
                tempu[index++] = c3[j];
                tempu[index++] = c4[j];
                tempu[index++] = c5[j];
                tempu[index++] = c6[j];
                tempu[index++] = c7[j];
                tempu[index++] = c8[j];
                tempu[index++] = c9[j];
                tempu[index++] = c10[j];
                tempu[index++] = c11[j];
                tempu[index++] = c12[j];
                tempu[index++] = c13[j];
                tempu[index++] = c14[j];
                tempu[index++] = c15[j];
                tempu[index++] = c16[j];
            }
        } else { // frame_size == FRAME_SIZE_SHORT
            rows = frame_size / mod;

            interleave_parity_bits(tempu, in);

            c1 = &tempv[0];
            c2 = &tempv[rows];
            c3 = &tempv[rows * 2];
            c4 = &tempv[rows * 3];
            c5 = &tempv[rows * 4];
            c6 = &tempv[rows * 5];
            c7 = &tempv[rows * 6];
            c8 = &tempv[rows * 7];

            twist_interleave_columns(tempv, tempu, rows, mod, twist256s);

            for (int j = 0; j < rows; j++) {
                tempu[index++] = c1[j];
                tempu[index++] = c2[j];
                tempu[index++] = c3[j];
                tempu[index++] = c4[j];
                tempu[index++] = c5[j];
                tempu[index++] = c6[j];
                tempu[index++] = c7[j];
                tempu[index++] = c8[j];
            }
        }
    }

    // tempu now has the input indices interleaved correctly, so save it
    memcpy(lookup_table, tempu, frame_size * sizeof(int));

    delete[] tempu;
    delete[] tempv;
}

/*
 * Our virtual destructor.
 */
dvbt2_interleaver_bb_impl::~dvbt2_interleaver_bb_impl() {}

void dvbt2_interleaver_bb_impl::forecast(int noutput_items,
                                         gr_vector_int& ninput_items_required)
{
    ninput_items_required[0] = noutput_items * mod;
}

int dvbt2_interleaver_bb_impl::general_work(int noutput_items,
                                            gr_vector_int& ninput_items,
                                            gr_vector_const_void_star& input_items,
                                            gr_vector_void_star& output_items)
{
    const unsigned char* in = (const unsigned char*)input_items[0];
    unsigned char* out = (unsigned char*)output_items[0];
    int consumed = 0;
    int produced = 0;
    int rows, offset, index, packed;
    unsigned int pack;
    const int* mux;

    switch (signal_constellation) {
    case MOD_QPSK:
        for (int i = 0; i < noutput_items; i += packed_items) {
            rows = frame_size / 2;
            if (code_rate == C1_3 || code_rate == C2_5) {
                for (int k = 0; k < nbch; k++) {
                    tempu[k] = *in++;
                }
                for (int t = 0; t < q_val; t++) {
                    for (int s = 0; s < 360; s++) {
                        tempu[nbch + (360 * t) + s] = in[(q_val * s) + t];
                    }
                }
                in = in + (q_val * 360);
                index = 0;
                for (int j = 0; j < rows; j++) {
                    out[produced] = tempu[index++] << 1;
                    out[produced++] |= tempu[index++];
                    consumed += 2;
                }
            } else {
                for (int j = 0; j < rows; j++) {
                    out[produced] = in[consumed++] << 1;
                    out[produced++] |= in[consumed++];
                }
            }
        }
        break;

    case MOD_16QAM:
        if (code_rate == C3_5 && frame_size == FRAME_SIZE_NORMAL) {
            mux = &mux16_35[0];
        } else if (code_rate == C1_3 && frame_size == FRAME_SIZE_SHORT) {
            mux = &mux16_13[0];
        } else if (code_rate == C2_5 && frame_size == FRAME_SIZE_SHORT) {
            mux = &mux16_25[0];
        } else {
            mux = &mux16[0];
        }
        packed = frame_size / (mod * 2);
        for (int i = 0; i < noutput_items; i += packed_items) {
            index = 0;
            for (int d = 0; d < packed; d++) {
                pack = 0;
                for (int e = 0; e < (mod * 2); e++) {
                    offset = mux[e];
                    pack |= in[lookup_table[index++]] << (((mod * 2) - 1) - offset);
                }
                out[produced++] = pack >> 4;
                out[produced++] = pack & 0xf;
            }
            consumed += frame_size;
            in += frame_size;
        }
        break;

    case MOD_64QAM:
        if (code_rate == C3_5 && frame_size == FRAME_SIZE_NORMAL) {
            mux = &mux64_35[0];
        } else if (code_rate == C1_3 && frame_size == FRAME_SIZE_SHORT) {
            mux = &mux64_13[0];
        } else if (code_rate == C2_5 && frame_size == FRAME_SIZE_SHORT) {
            mux = &mux64_25[0];
        } else {
            mux = &mux64[0];
        }
        packed = frame_size / (mod * 2);
        for (int i = 0; i < noutput_items; i += packed_items) {
            index = 0;
            for (int d = 0; d < packed; d++) {
                pack = 0;
                for (int e = 0; e < (mod * 2); e++) {
                    offset = mux[e];
                    pack |= in[lookup_table[index++]] << (((mod * 2) - 1) - offset);
                }
                out[produced++] = pack >> 6;
                out[produced++] = pack & 0x3f;
            }
            consumed += frame_size;
            in += frame_size;
        }
        break;

    case MOD_256QAM:
        if (frame_size == FRAME_SIZE_NORMAL) {
            if (code_rate == C3_5) {
                mux = &mux256_35[0];
            } else if (code_rate == C2_3) {
                mux = &mux256_23[0];
            } else {
                mux = &mux256[0];
            }
            packed = frame_size / (mod * 2);
            for (int i = 0; i < noutput_items; i += packed_items) {
                index = 0;
                for (int d = 0; d < packed; d++) {
                    pack = 0;
                    for (int e = 0; e < (mod * 2); e++) {
                        offset = mux[e];
                        pack |= in[lookup_table[index++]] << (((mod * 2) - 1) - offset);
                    }
                    out[produced++] = pack >> 8;
                    out[produced++] = pack & 0xff;
                }
                consumed += frame_size;
                in += frame_size;
            }
        }

        else { // frame_size = FRAME_SIZE_SHORT
            if (code_rate == C1_3) {
                mux = &mux256s_13[0];
            } else if (code_rate == C2_5) {
                mux = &mux256s_25[0];
            } else {
                mux = &mux256s[0];
            }
            packed = frame_size / mod;
            for (int i = 0; i < noutput_items; i += packed_items) {
                index = 0;
                for (int d = 0; d < packed; d++) {
                    pack = 0;
                    for (int e = 0; e < mod; e++) {
                        offset = mux[e];
                        pack |= in[lookup_table[index++]] << ((mod - 1) - offset);
                    }
                    out[produced++] = pack & 0xff;
                }
                consumed += frame_size;
                in += frame_size;
            }
        }
        break;
    }

    // Tell runtime system how many input items we consumed on
    // each input stream.
    consume_each(consumed);

    // Tell runtime system how many output items we produced.
    return noutput_items;
}

const int dvbt2_interleaver_bb_impl::twist16n[8] = { 0, 0, 2, 4, 4, 5, 7, 7 };

const int dvbt2_interleaver_bb_impl::twist64n[12] = {
    0, 0, 2, 2, 3, 4, 4, 5, 5, 7, 8, 9
};

const int dvbt2_interleaver_bb_impl::twist256n[16] = { 0,  2,  2,  2,  2,  3,  7,  15,
                                                       16, 20, 22, 22, 27, 27, 28, 32 };

const int dvbt2_interleaver_bb_impl::twist16s[8] = { 0, 0, 0, 1, 7, 20, 20, 21 };

const int dvbt2_interleaver_bb_impl::twist64s[12] = {
    0, 0, 0, 2, 2, 2, 3, 3, 3, 6, 7, 7
};

const int dvbt2_interleaver_bb_impl::twist256s[8] = { 0, 0, 0, 1, 7, 20, 20, 21 };

const int dvbt2_interleaver_bb_impl::mux16[8] = { 7, 1, 4, 2, 5, 3, 6, 0 };

const int dvbt2_interleaver_bb_impl::mux64[12] = { 11, 7, 3, 10, 6, 2, 9, 5, 1, 8, 4, 0 };

const int dvbt2_interleaver_bb_impl::mux256[16] = { 15, 1, 13, 3, 8,  11, 9,  5,
                                                    10, 6, 4,  7, 12, 2,  14, 0 };

const int dvbt2_interleaver_bb_impl::mux16_35[8] = { 0, 5, 1, 2, 4, 7, 3, 6 };

const int dvbt2_interleaver_bb_impl::mux16_13[8] = { 6, 0, 3, 4, 5, 2, 1, 7 };

const int dvbt2_interleaver_bb_impl::mux16_25[8] = { 7, 5, 4, 0, 3, 1, 2, 6 };

const int dvbt2_interleaver_bb_impl::mux64_35[12] = {
    2, 7, 6, 9, 0, 3, 1, 8, 4, 11, 5, 10
};

const int dvbt2_interleaver_bb_impl::mux64_13[12] = {
    4, 2, 0, 5, 6, 1, 3, 7, 8, 9, 10, 11
};

const int dvbt2_interleaver_bb_impl::mux64_25[12] = {
    4, 0, 1, 6, 2, 3, 5, 8, 7, 10, 9, 11
};

const int dvbt2_interleaver_bb_impl::mux256_35[16] = { 2,  11, 3, 4,  0, 9,  1, 8,
                                                       10, 13, 7, 14, 6, 15, 5, 12 };

const int dvbt2_interleaver_bb_impl::mux256_23[16] = { 7,  2,  9,  0, 4, 6,  13, 3,
                                                       14, 10, 15, 5, 8, 12, 11, 1 };

const int dvbt2_interleaver_bb_impl::mux256s[8] = { 7, 3, 1, 5, 2, 6, 4, 0 };

const int dvbt2_interleaver_bb_impl::mux256s_13[8] = { 4, 0, 1, 2, 5, 3, 6, 7 };

const int dvbt2_interleaver_bb_impl::mux256s_25[8] = { 4, 0, 5, 1, 2, 3, 6, 7 };

} /* namespace dtv */
} /* namespace gr */